GitLab

PSCI requires a core to turn itself off, which we can't do properly by
just executing an algorithm on that very core. As a consequence we just
put a core into WFI on CPU_OFF right now.
To fix this let's task the "arisc" management processor (an OpenRISC
core) with that task of asserting reset and turning off the core's power
domain. We use a handcrafted sequence of OpenRISC instructions to
achieve this, and hand this data over to the new sunxi_execute_arisc_code()
routine.
The commented source code for this routine is provided in a separate file,
but the ATF code contains the already encoded instructions as data.
The H6 uses the same algorithm, but differs in the MMIO addresses, so
provide a SoC (family) specific copy of that code.
Signed-off-by: Andre Przywara <andre.przywara@arm.com>

The more recent Allwinner SoCs contain an OpenRISC management
controller (called arisc or CPUS), which shares the bus with the ARM cores,
but runs on a separate power domain. This is meant to handle power
management with the ARM cores off.
There are efforts to run sophisticated firmware on that core
(communicating via SCPI with the ARM world), but for now can use it for
the rather simple task of helping to turn the ARM cores off. As this
cannot be done by ARM code itself (because execution stops at the
first of the three required steps), we can offload some instructions to
this management processor.
This introduces a helper function to hand over a bunch of instructions
and triggers execution. We introduce a bakery lock to avoid two cores
trying to use that (single) arisc core. The arisc code is expected to
put itself into reset after is has finished execution.
Signed-off-by: Andre Przywara <andre.przywara@arm.com>

There are reports that activating the DC1SW before certain other
regulators leads to the PMIC overheating and consequently shutting down.
To avoid this situation, delay the activation of the DC1SW line until
the very end, so those other lines are always activated earlier.
Signed-off-by: Andre Przywara <andre.przywara@arm.com>

Based on the just introduced PMIC FDT framework, we check the DT for more
voltage rails that need to be setup early:
- DCDC1 is typically the main board power rail, used for I/O pins, for
instance. The PMIC's default is 3.0V, but 3.3V is what most boards use,
so this needs to be adjusted as soon as possible.
- DCDC5 is supposed to be connected to the DRAM. The AXP has some
configurable reset voltage, but some boards get that wrong, so we better
set up this here to avoid over- or under-volting.
- DLDO1,2,3 and FLDO1 mostly drive some graphics related IP, some boards
need this to be up to enable HDMI or the LCD screen, so we get screen
output in U-Boot.

To get the right setup, but still being flexible, we query the DT for
the required voltage and whether that regulator is actually used. That
gives us some robust default setup U-Boot is happy with.
Signed-off-by: Andre Przywara <andre.przywara@arm.com>

Now that we have a pointer to the device tree blob, let's use that to
do some initial setup of the PMIC:
- We scan the DT for the compatible string to find the PMIC node.
- We switch the N_VBUSEN pin if the DT property tells us so.
- We scan over all regulator subnodes, and switch DC1SW if there is at
least one other node referencing it (judging by the existence of a
phandle property in that subnode).
This is just the first part of the setup, a follow up patch will setup
voltages.
Signed-off-by: Andre Przywara <andre.przywara@arm.com>

For Allwinner boards we now use some heuritistics to find a preloaded
.dtb file.

Pass this address on to the PMIC setup routine, so that it can use the
information contained therein to setup some initial power rails.
Signed-off-by: Andre Przywara <andre.przywara@arm.com>

The initial PMIC setup for the Allwinner platform is quite board
specific, and used to be guarded by reading the .dtb stub *name* from the
SPL image in the legacy ATF port. This doesn't scale particularly well,
and requires constant maintainance.
Instead having the actual .dtb available would be much better, as the PMIC
setup requirements could be read from there directly.
The only available BL33 for Allwinner platforms so far is U-Boot, and
fortunately U-Boot comes with the full featured .dtb, appended to the
end of the U-Boot image.

Introduce some code that scans the beginning of the BL33 image to look
for the load address, which is followed by the image size. Adding those
two values together gives us the end of the image and thus the .dtb
address. Verify that this heuristic is valid by sanitising some values
and checking the DTB magic.

Print out the DTB address and the model name, if specified in the root
node.
Signed-off-by: Andre Przywara <andre.przywara@arm.com>

Boards with the Allwinner A64 SoC are mostly paired with an AXP803 PMIC,
which allows to programmatically power down the board.

Use the newly introduced RSB driver to detect and program the PMIC on
boot, then later to turn off the main voltage rails when receiving a
PSCI SYSTEM_POWER_OFF command.
Signed-off-by: Andre Przywara <andre.przywara@arm.com>

In the H6 platform code there is a routine to do the platform
initialisation of the R_I2C controller. We will need a very similar
setup routine to initialise the RSB controller on the A64.

Move this code to sunxi_common.c and generalise it to support all SoCs
and also to cover the related RSB bus.
Signed-off-by: Andre Przywara <andre.przywara@arm.com>

Allwinner produces reference board designs, which apparently most board
vendors copy from. So every H5 board I checked uses regulators which are
controlled by the same PortL GPIO pins to power the ARM CPU cores, the
DRAM and the I/O ports.
Add a SoC specific power down routine, which turns those regulators off
when ATF detects running on an H5 SoC and the rich OS triggers a
SYSTEM_POWEROFF PSCI call.

NOTE: It sounds very tempting to turn the CPU power off, but this is not
working as expected, instead the system is rebooting. Most probably this
is due to VCC-SYS also being controlled by the same GPIO line, and
turning this off requires an elaborate and not fully understood setup.
Apparently not even Allwinner reference code is turning this regulator
off. So for now we refrain to pulling down PL8, the power consumption is
quite low anyway, so we are as close to poweroff as reasonably possible.
Many thanks to Samuel for doing some research on that topic.
Signed-off-by: Andre Przywara <andre.przywara@arm.com>

Many boards without a dedicated PMIC contain simple regulators, which
can be controlled via GPIO pins.

To later allow turning them off easily, introduce a simple function to
configure a given pin as a GPIO out pin and set it to the desired level.
Signed-off-by: Andre Przywara <andre.przywara@arm.com>

So far we have a sunxi_private.h header file in the common code directory.
This holds the prototypes of various functions we share in *common*
code. However we will need some of those in the platform specific code
parts as well, and want to introduce new functions shared across the
whole platform port.

So move the sunxi_private.h file into the common/include directory, so
that it becomes visible to all parts of the platform code.
Fix up the existing #includes and add missing ones, also add the
sunxi_read_soc_id() prototype here.

This will be used in follow up patches.
Signed-off-by: Andre Przywara <andre.przywara@arm.com>

Some boards don't have a PMIC, so they can't easily turn their power
off. To cover those boards anyway, let's turn off as many devices and
clocks as possible, so that the power consumption is reduced. Then
halt the last core, as before.
This will later be extended with proper PMIC support for supported
boards.
Signed-off-by: Andre Przywara <andre.przywara@arm.com>

In the BL31 platform setup we read the Allwinner SoC ID to identify the
chip and print its name.
In addition to that we will need to differentiate the power setup
between the SoCs, to pass on the SoC ID to the PMIC setup routine.
Signed-off-by: Andre Przywara <andre.przywara@arm.com>

We will soon make more use of the Allwinner SoC ID, to differentiate the
platform setup.
Introduce definitions to avoid dealing with magic numbers and make the
code more readable.
Signed-off-by: Andre Przywara <andre.przywara@arm.com>

The SRAM in the Allwinner H6 SoC starts at 0x2000, with the last part
ending at 0x117fff (although with gaps in between).
So SUNXI_SRAM_SIZE should be 0xf8000, not 0x98000.

Fix this to map the arisc exception vector area, which we will need
shortly.
Signed-off-by: Andre Przywara <andre.przywara@arm.com>

According to the documentation, platforms may choose to trade memory
footprint for performance (and elegancy) by not providing a separately
mapped coherent page.

Since a debug build is getting close to the SRAM size limit already, this
allows us to save about 3.5KB of BSS and have some room for future
enhancements.
Signed-off-by: Andre Przywara <andre.przywara@arm.com>

At the moment we map as much of the DRAM into EL3 as possible, however
we actually don't use it. The only exception is the secure DRAM for
BL32 (if that is configured).

To decrease the memory footprint of ATF, we save on some page tables by
reducing the memory mapping to the actually required regions: SRAM, device
MMIO, secure DRAM and U-Boot (to be used later).
This introduces a non-identity mapping for the DRAM regions.
Signed-off-by: Andre Przywara <andre.przywara@arm.com>

For the two different platforms we support in the Allwinner port we
mostly rely on header files covering the differences. This leads to the
platform.mk files in the respective directories to be almost identical.

To avoid further divergence and make sure that one platform doesn't
break accidentally, let's create a shared allwinner-common.mk file and
include that from the platform directory.
Signed-off-by: Andre Przywara <andre.przywara@arm.com>

Change-Id: I206478597dd9855d3fe1577e7e2c0fe6d2af1cc5
Tested-by: Andre Przywara <andre.przywara@arm.com>
Signed-off-by: Antonio Nino Diaz <antonio.ninodiaz@arm.com>

At the moment we have two I2C stub drivers (for the Allwinner and the
Marvell platform), which #include the actual .c driver file.
Change this into the more usual design, by renaming and moving the stub
drivers into platform specific header files and including these from the
actual driver file. The platform specific include directories make sure
the driver picks up the right header automatically.
Signed-off-by: Andre Przywara <andre.przywara@arm.com>

Even though we initialise the platform part and the I2C controller
itself at boot time, we actually only access the bus on power down.
Meanwhile a rich OS might have configured the I2C pins differently or
even disabled the controller.
So repeat the platform setup and controller initialisation just before
we actually access the bus to power off the system. This is safe,
because at this point the rich OS should no longer be running.
Signed-off-by: Andre Przywara <andre.przywara@arm.com>

Drop the unnecessary check for the I2C pins being already configured as
I2C pins (we actually don't care).
Also avoid resetting *every* peripheral that is covered by the PRCM reset
controller, instead just clear the one line connected to the I2C controller.
Signed-off-by: Andre Przywara <andre.przywara@arm.com>

The AXP805 PMIC used with H6 is capable of shutting down the system.

Add support for using it to shut down the system power.

The original placeholder power off code is moved to A64 code, as it's
still TODO to implement PMIC operations for A64.
Signed-off-by: Icenowy Zheng <icenowy@aosc.io>

The OTT reference design of Allwinner H6 SoC uses an X-Powers AXP805
PMIC.

Add initial code for it.

Currently it's only detected.
Signed-off-by: Icenowy Zheng <icenowy@aosc.io>

As the ATF may need to do some power initialization on Allwinner
platform with AXP PMICs, call the PMIC setup code in BL31.

Stub of PMIC setup code is added, to prevent undefined reference.
Signed-off-by: Icenowy Zheng <icenowy@aosc.io>

Also change header guards to fix defects of MISRA C-2012 Rule 21.1.

Change-Id: Ied0d4b0e557ef6119ab669d106d2ac5d99620c57
Acked-by: Sumit Garg <sumit.garg@linaro.org>
Acked-by: Anson Huang <Anson.Huang@nxp.com>
Signed-off-by: Antonio Nino Diaz <antonio.ninodiaz@arm.com>

The H6 is Allwinner's most recent SoC. It shares most peripherals with the
other ARMv8 Allwinner SoCs (A64/H5), but has a completely different memory
map.

Introduce a separate platform target, which includes a different header
file to cater for the address differences. Also add the new build target
to the documentation.

The new ATF platform name is "sun50i_h6".
Signed-off-by: Andre Przywara <andre.przywara@arm.com>

So far we already support booting on two different SoCs, and we will
shortly add a third, so add some code to determine the current SoC type.
This can be later used to runtime detect certain properties.

Also print the SoC name to the console, to give valuable debug information.
Signed-off-by: Andre Przywara <andre.przywara@arm.com>

There is nothing we need from the BootROM area, so we also don't need
to map it in EL3.

Remove the mapping and reduce the number of MMAP regions by one.
Reported-by: Samuel Holland <samuel@sholland.org>
Signed-off-by: Andre Przywara <andre.przywara@arm.com>

The DRAM controller supports up to 4GB of DRAM, and there are actually
boards out there where we can use at least 3GB of this.

Relax the PSCI entry point check, to be not restricted to 2GB of DRAM.
Signed-off-by: Andre Przywara <andre.przywara@arm.com>

The "#ifdef SUNXI_SPC_BASE" guard was meant to allow the build on SoCs
without a Secure Peripherals Controller, so that we skip that part of
the security setup. But in the current position this will trigger a
warning about an unused variable.

Simply move the guard one line up to cover the variable as well.
Signed-off-by: Andre Przywara <andre.przywara@arm.com>

The "INFO" output in sunxi_cpu_ops.c is quite verbose, so make this more
obvious by changing the log level to "VERBOSE" and so avoiding it to
be printed in a normal (even debug) build.
Reported-by: Samuel Holland <samuel@sholland.org>
Signed-off-by: Andre Przywara <andre.przywara@arm.com>

The relative VER_REG *offset* is the same across all known SoCs, so we
can define this offset near it's user.

Remove it from the memory map.
Reported-by: Samuel Holland <samuel@sholland.org>
Signed-off-by: Andre Przywara <andre.przywara@arm.com>

Some code in sunxi_common.c requires symbols defined in sunxi_private.h,
so add the header to that file.
It was included via another header before, but let's make this explicit.
Reported-by: Samuel Holland <samuel@sholland.org>
Signed-off-by: Andre Przywara <andre.przywara@arm.com>

This patch is an attempt to run Trusted OS (OP-TEE OS being one of them) along
side BL31 image.

ATF supports multiple SPD's that can take dispatcher name (opteed for OP-TEE OS)
as an input using the 'SPD=<dispatcher name>' option during bl31 build.
Signed-off-by: Amit Singh Tomar <amittomer25@gmail.com>

This patch is an attempt to run Trusted OS (OP-TEE OS being one of them) along
side BL31 image.

ATF supports multiple SPD's that can take dispatcher name (opteed for OP-TEE OS)
as an input using the 'SPD=<dispatcher name>' option during bl31 build.
Signed-off-by: Amit Singh Tomar <amittomer25@gmail.com>

Some peripherals are TrustZone aware, so they need to be configured to
be accessible from non-secure world, as we don't need any of them being
exclusive to the secure world.
This affects some clocks, DMA channels and the Secure Peripheral
Controller (SPC). The latter controls access to most devices, but is not
active unless booting with the secure boot fuse burnt.
Signed-off-by: Andre Przywara <andre.przywara@arm.com>

The reset vector entry point is preserved across CPU resets, so it only
needs to be set once at boot.

Hotplugged CPUs are not actually powered down, but are put in a wfi with
the GIC disconnected.

With this commit, Linux is able to enable, hotplug and use all four CPUs.
Signed-off-by: Samuel Holland <samuel@sholland.org>
Signed-off-by: Andre Przywara <andre.przywara@arm.com>

sun50i_cpu_on will be used by the PSCI implementation to initialize
secondary cores for SMP. Unfortunately, sun50i_cpu_off is not usable by
PSCI directly, because it is not possible for a CPU to use this function
to power itself down. Power cannot be shut off until the outputs are
clamped, and MMIO does not work once the outputs are clamped.

But at least CPU0 can shutdown the other cores early in the BL31 boot
process and before shutting down the system.
Signed-off-by: Samuel Holland <samuel@sholland.org>
Signed-off-by: Andre Przywara <andre.przywara@arm.com>